verificationists want to treat as verifiable and exclude those they do not. Here is a simple
example of the latter. We can take any nonobservational statement, for example, “God is
perfectly good,” and make it subject to empirical test by conjoining it with a hypothetical
statement like “If God is perfectly good, then it will rain tomorrow here.” This
conjunction implies “It will rain tomorrow here,” and this makes an observation of the
weather have a bearing on the justification of the theological statement. No doubt, it
would be absurd to accept this bridge principle. But bridge principles in science often
have no antecedent plausibility. And despite the expenditure of a lot of effort, no one has
been able to come up with a plausible criterion of acceptability for bridge principles that
will let in accepted scientific examples and keep out theological and metaphysical
examples.
Here is another indication of what is wrong with verificationism. In the history of
science, hypotheses, for example, the atomic hypothesis concerning the constitution of
matter, were originally put forward without anyone as yet having any idea as to how they
could be empirically tested. Eventually the atomic hypothesis was brought into effective
connection with empirical tests. But unless the hypothesis was understandable as a
factual claim at the earlier stages, those developments would have been impossible.
If one is convinced, despite the criticisms just mounted, that no utterances about God, as
construed in developed theistic religion, are factually meaningful, how will one construe
them? There are a number of alternatives. The simplest one is to avoid the necessity of
any reconception by ignoring them altogether or, in Hume's memorable phrase,
consigning them to the flames. But if one is sufficiently motivated to retain God-talk,
there are a number of ways to do so while avoiding any reference to a transcendent deity.
These can be divided into two main groups. One seeks to preserve the statemental
character by giving a purely natural-world meaning to God-talk. The other chooses to
interpret putatively
end p.223
statemental talk about God as expressive of feelings, attitudes, commitments, and the
like.
Here are two examples of the first alternative. The American theologian Henry Nelson
Wieman defines “God” in naturalistic terms as “that interaction between individuals,
groups, and ages which generates and promotes the greatest possible mutuality of good”
(Wieman, McIntosh, and Otto 1932, 13). This preserves the beneficence of God, but the
personal being is completely lost. In defense of his suggestion, Wieman has this to say:
“Can men pray to an interaction? Yes, that is what they always pray to, under any
concept of God. Can men love an interaction? Yes, that is what they always love. When I
love Mr. Jones, it is not Mr. Jones in the abstract, but the fellowship of Mr. Jones.
Fellowship is a kind of interaction” (ibid., 17). Mr. Jones would no doubt be disappointed
to learn that what was loved was not himself but rather fellowship with him.
Another naturalistic reinterpretation of theistic talk is found in the English biologist
Julian Huxley's book Religion without Revelation (1957). He identifies God the Father
with the forces of nonhuman nature (the “creator”), God the Holy Spirit as the ideals for
which men are striving (at their best), and God the Son as human nature itself, which is,
more or less, utilizing the forces of nature in the pursuit of those ideals. Thus, he gives us